PROJECT SUMMARY/ABSTRACT Candidate: Dr. Adel El Boueiz is a pulmonary and critical care physician-scientist completing a period of T32- funded support at the Channing Division of Network Medicine (CDNM) and Harvard Medical School (HMS). He received a Master's of Medical Science in Biomedical Informatics from HMS in May 2016. He will be promoted to Instructor of Medicine at the CDNM and HMS on July 1, 2017. His principal research interests are the genetic epidemiology of chronic obstructive pulmonary disease (COPD) and the translation of genomic discoveries into clinical practice and public health. His long-term goal is to be an independent investigator with expertise in imaging phenotyping, genomics, and predictive analytics of the regional heterogeneity of the various aspects of COPD (emphysema, airway disease, and pulmonary vascular remodeling). Environment: Dr. El Boueiz will continue to pursue his research and career development in the rich and multidisciplinary environment of the CDNM and the Brigham and Women's Hospital Applied Chest Imaging Lab (ACIL). He will be mentored by Drs. Edwin K. Silverman, Peter J. Castaldi, and Ral San Jos Estpar, leaders in the field of COPD quantitative imaging, genetic epidemiology, and predictive analytics with excellent track records of mentoring young investigators towards independent research careers. His career development will also be overseen by an advisory committee with expertise related to key areas of his proposal. Research: COPD is a major cause of morbidity and mortality that is of increasing public health importance. COPD is a heterogeneous disease and this heterogeneity complicates the identification of the predictors of disease progression and consequently, the development of effective therapies. Emphysema distribution is an important COPD-related phenotype that emerged as a strong predictor of the response to lung volume reduction procedures. Despite the availability of advanced texture-based CT quantification methods, global threshold-based quantitative metrics have to date been the cornerstone for the radiological characterization of emphysema distribution with inability to differentiate centrilobular, panlobular, and paraseptal emphysema patterns. In this project, we will apply a texture-based CT quantification method to discover novel imaging biomarkers of the regional heterogeneity of centrilobular, panlobular, and paraseptal emphysema in a large cohort of well-characterized smokers and identify their genetic determinants using whole genome sequencing and integrative genomics analyses. The results will be considered for inclusion along with other rich phenotypic and imaging data in COPD disease progression machine learning predictive models. Relevance: Through improved radiographic phenotyping of emphysema distribution, better understanding of disease pathobiology, and more accurate prediction of disease progression, the proposed work will open new avenues of investigation for the development of personalized and improved COPD therapeutic strategies.